arX

iv:0

812.

0793

v1 [

astr

o-ph

] 3

Dec

200

8

Section on Extragalactic Science Topics of the White Paper on the

Status and Future of Ground-Based TeV Gamma-Ray Astronomy

H. Krawczynski (Wash. University in St. Louis, Physics Department and McDonnellCenter for the Space Sciences), A. M. Atoyan (Montreal University), M. Beilicke (Wash.

University in St. Louis, Physics Department and McDonnell Center for the SpaceSciences), R. Blandford (Kavli Institute for Particle Astrophysics and Cosmology, StanfordLinear Accelerator Center), M. Boettcher (Ohio University), J. Buckley (Wash. University

in St. Louis, Physics Department and McDonnell Center for the Space Sciences),

A. Carraminana (Instituto Nacional de Astrofısica, Optica y Electronica, Mexico),P. Coppi (Yale University), C. Dermer (U.S. Naval Research Laboratory), B. Dingus (LosAlamos), E. Dwek (NASA Goddard Space Flight Center), A. Falcone (Pennsylvania StateUniversity), S. Fegan (University of California, Los Angeles, now at: Institut National de laPhysique Nucleaire), J. Finley (Purdue University), S. Funk (Kavli Institute for ParticleAstrophysics and Cosmology, Stanford Linear Accelerator Center), M. Georganopoulos(University of Maryland, NASA Goddard Space Flight Center), J. Holder (University of

Delaware), D. Horan (Argonne National Laboratory, now at: Institut National de laPhysique Nucleaire), T. Jones (University of Minnesota), I. Jung (Wash. University in St.

Louis, Physics Department and McDonnell Center for the Space Sciences, now at:Universitat Erlangen-Nurnberg), P. Kaaret (The University of Iowa), J. Katz (Wash.

University in St. Louis, Physics Department and McDonnell Center for the SpaceSciences), F. Krennrich (Iowa State University), S. LeBohec (University of Utah),

J. McEnery (NASA Goddard Space Flight Center), R. Mukherjee (Columbia University),R. Ong (University of California, Los Angeles), E. Perlman (Flordia Institute of

Technology), M. Pohl (Iowa State University), S. Ritz (NASA Goddard Space FlightCenter), J. Ryan (University of New Hampshire), G. Sinnis (Los Alamos NationalLaboratory), V. Vassiliev (University of California, Los Angeles), M. Urry (Yale

University), T. Weekes (Smithsonian Astrophysical Observatory).

ABSTRACT

This is a report on the findings of the extragalactic science working group for the white paperon the status and future of TeV gamma-ray astronomy. The white paper was commissioned by theAmerican Physical Society, and the full white paper can be found on astro-ph (arXiv:0810.0444).This detailed section discusses extragalactic science topics including active galactic nuclei, cosmicray acceleration in galaxies, galaxy clusters and large scale structure formation shocks, and thestudy of the extragalactic infrared and optical background radiation. The scientific potentialof ground based gamma-ray observations of Gamma-Ray Bursts and dark matter annihilationradiation is covered in other sections of the white paper.

1

1. Introduction

A next-generation gamma-ray experiment willmake extragalactic discoveries of profound impor-tance. Topics to which gamma-ray observationscan make unique contributions are the following:(i) the environment and growth of SupermassiveBlack Holes; (ii) the acceleration of cosmic raysin other galaxies; (iii) the largest particle accel-erators in the Universe, including radio galaxies,galaxy clusters, and large scale structure forma-tion shocks; (iv) study of the integrated electro-magnetic luminosity of the Universe and inter-galactic magnetic field strengths through processesincluding pair creation of TeV gamma rays inter-acting with infrared photons from the Extragalac-tic Background Light (EBL).

The scientific potential of ground based observa-tions of gamma-ray bursts and γ-rays from darkmatter annihilation processes is described in othersections of the white paper (1).

2. Gamma-ray observations of supermas-

sive black holes

Supermassive black holes (SMBH) have masses be-tween a million and several billion solar massesand exist at the centers of galaxies. Some SMBHs,called Active Galactic Nuclei (AGN) are strongemitters of electromagnetic radiation. Observa-tions with the EGRET Energetic Gamma-Ray

Experiment Telescope on board of the ComptonGamma-Ray Observatory (CGRO) revealed thata certain class of AGN known as blazars arepowerful and variable emitters, not just at ra-dio through optical wavelengths, but also at ≥100MeV gamma-ray energies (6). EGRET largelydetected quasars, the most powerful blazars inthe Universe. Observations with ground-basedCherenkov telescopes showed that blazars emiteven at TeV energies (7). In the meantime, morethan twenty blazars have now been identified assources of >200 GeV gamma rays with redshiftsranging from 0.031 (Mrk 421) (7) to 0.536 (3C 279)(72) 1. Most TeV bright sources are BL Lac typeobjects, the low power counterparts of the quasarsdetected by EGRET. The MeV to TeV gamma-ray emission from blazars is commonly thought tooriginate from highly relativistic collimated out-flows (jets) from mass accreting SMBHs that point

1Up-to-date lists of TeV γ-ray sources can be foundat the web-sites: http: //tevcat.uchicago.edu and http://www.mpp.mpg.de/ ∼rwagner/sources/.

at the observer (4; 5). The only gamma-ray emit-ting AGN detected to date that are not blazarsare the radio galaxies Centaurus A (2) and M87(3). The observation of blazars in the gamma-rayband has had a major impact on our understand-ing of these sources. The observation of rapid fluxvariability on time scales of minutes together withhigh gamma-ray and optical fluxes (12; 66) im-plies that the accreting black hole gives rise toan extremely relativistic jet-outflow with a bulkLorentz factor exceeding 10, most likely even inthe range between 10 and 50 (67; 68). Gamma-ray observations thus enable us to study plasmawhich moves with ≥99.98% of the speed of light.Simultaneous broadband multiwavelength obser-vations of blazars have revealed a pronounced cor-relation of the X-ray and TeV gamma-ray fluxes(13; 14; 17; 8). The X-ray/TeV flux correlation(see Fig. 1) suggests that the emitting particlesare electrons radiating synchrotron emission in theradio to X-ray band and inverse Compton emissionin the gamma-ray band.

Blazars are expected to be the most copiousextragalactic sources detected by ground-basedIACT arrays like VERITAS and by the satelliteborne gamma-ray telescope Fermi. For extremelystrong sources, IACT arrays will be able to trackGeV/TeV fluxes on time scales of seconds andGeV/TeV energy spectra on time scales of a fewminutes. Resolving the spectral variability duringindividual strong flares in the X-ray and gamma-ray bands should lead to the unambiguous identi-fication of the emission mechanism. The presentgeneration of IACTs will be able to track spec-tral variations only for a very small number ofsources and only during extreme flares. The next-generation gamma-ray experiments will be able todo such studies for a large number of sources ona routine basis. Sampling the temporal variationof broadband energy spectra from a few tens ofGeV to several TeV will allow us to use blazarsas precision laboratories to study particle accel-eration and turbulence in astrophysical plasmas,and to determine the physical parameters describ-ing a range of different AGN. The observations ofblazars hold the promise to reveal details aboutthe inner workings of AGN jets. Obtaining realis-tic estimates of the power in the jet, and the jetmedium will furthermore constrain the origin ofthe jet and the nature of the accretion flow.

Recently, spectracular results have been obtainedby combining monitoring VLBA, X-ray and TeVγ-ray observations. This combination has the po-

2

γ-r

ay[C

rab]

X-r

ay[c

ts/s

/PC

U]

optic

al[V

mag

.]

Fig. 1.—: Results from 2001 Rossi X-ray Timing Explorer (RXTE) 2-4 keV X-ray and Whipple (full symbols) and HEGRA(open symbols) gamma-ray observations of Mrk 421 in the year 2001 (8). The X-ray/gamma-ray fluxes seem to be correlated.However, the interpretation of the data is hampered by the sparse coverage at TeV gamma rays.

tential to pinpoint the origin of the high energyemission based on the high resolution radio im-ages, and thus to directly confirm or to refute mod-els of jet formation. For example, radio VLBA,optical polarimetry, X-ray and TeV γ-ray observa-tions of the source BL Lac seem to indicate thata plasma blob first detected with the VLBA sub-sequently produces an X-ray, an optical and a γ-ray flare (70). A swing of the optical polarizationseems to bolster the case for a helical magneticfield as predicted by magnetic models of jet forma-tion and acceleration. Presently such observationsare extremely difficult as the current instrumentscan detect sources like M 87, BL Lac, W Comonly in long observations or during extreme flares.Next-generation γ-ray instruments will allow us tostudy the correlation of fast TeV flares and radiofeatures on a routine basis.

In addition to ground-based radio to optical cov-erage, several new opportunities might open up

within the next decade. The Space Interferome-try Mission (SIM) will be able to image emergingplasma blobs with sub milli-arcsec angular resolu-tion (15). The center may be located with an ac-curacy of a few micro-arcsec. For a nearby blazarat z=0.03, 1 milli-arcsec corresponds to a pro-jected distance of 0.6 pc. The SIM observationscould thus image the blobs that give rise to theflares detected in the gamma-ray regime. Joint X-ray/radio interferometry observations already givesome tentative evidence for the emergence of radioblobs correlated with X-ray flares. If a Black HoleFinder Probe like the Energetic X-ray ImagingSpace Telescope (EXIST) (16) will be launched,it would provide reliable all-sky, broad-bandwidth(0.5-600 keV), and high-sensitivity X-ray coveragefor all blazars in the sky. EXIST’s full-sky sensi-tivity would be 2 × 10−12 ergs cm−2 s−1 for 1month of integration. For bright sources, EXISTcould measure not only flux variations but also the

3

polarization of hard X-rays. Opportunities arisingfrom neutrino coverage will be described below.

At the time of writing this white paper, the Fermigamma-ray telescope is in the process of detect-ing a few thousand blazars. The source samplewill make it possible to study the redshift depen-dent luminosity function of blazars, although theidentification of sources with optical counterpartsmay be difficult for the weaker sources of the sam-ple, owing to Fermi’s limited angular resolution.Another important task for the next-generationinstrument will be to improve on the Fermi lo-calization accuracies, and thus to identify a largenumber of the weaker Fermi sources.

Independent constraints on the jet power, kine-matics, and emission processes can be derivedfrom GeV-TeV observations of the large scale (upto hundreds of kpc) jets recently detected byChandra. Although such large scale jets will notbe spatially resolved, the fact that the gamma-ray emission from the quasar core is highly vari-able permits us to set upper limits to the steadyGeV-TeV large scale jet emission (18). In thecase of the relatively nearby 3C 273, for exam-ple, the electrons that produce the large scale jetIR emission will also produce a flat GeV compo-nent. The fact that this emission is weaker thanthe EGRET upper limit constrains the Dopplerfactor of the large scale jets to less than 12, a valuethat can be pushed down to 5 with Fermi observa-tions. Such low values of delta have implicationson the nature of the large scale jet X-ray emissionobserved by Chandra. In particular, they disfa-vor models in which the X-ray emission is inverseCompton scattering of the cosmic microwave back-ground (CMB), because the jet power required in-creases beyond the so-called Eddington luminos-ity, thought by many to be the maximum lumi-nosity that can be channeled continuously in a jet.A synchrotron interpretation for the X-ray emis-sion, requiring significantly less jet power, pos-tulates a population of multi-TeV electrons thatwill unavoidably up-scatter the CMB to TeV en-ergies. The existing 3C 273 shallow HESS upperlimit constrains the synchrotron interpretation toDoppler factors less than 10. Combining deepTeV observations with a next-generation exper-iment with Fermi observations holds the promiseof confirming or refuting the synchrotron interpre-tation and constraining the jet power.

Whereas the X-ray/gamma-ray correlation favorsleptonic models with electrons as the emitters ofthe observed gamma-ray emission, hadronic mod-

els are not ruled out. In the latter case, thehigh-energy component is synchrotron emission,either from extremely high-energy (EHE) protons(31; 32; 33), or from secondary e+/e− resultingfrom synchrotron and pair-creation cascades ini-tiated by EHE protons (34) or high-energy elec-trons or photons (35; 36; 37; 38). If blazars in-deed accelerate UHE protons, it might even bepossible to correlate their TeV gamma-ray emis-sion with their flux of high-energy neutrinos de-tected by the IceCube detector (39). The highsensitivity of a next-generation ground-based ex-periment would be ideally suited to perform suchmulti-messenger studies.

Although most observations can be explained withthe emission of high-energy particles that are ac-celerated in the jets of AGN , the observations donot exclude that the emitting particles are acceler-ated closer to the black hole. If the magnetic fieldin the black hole magnetosphere has a poloidal netcomponent on the order of B100 = 100 G, both thespinning black hole (20) and the accretion disk(21; 19) will produce strong electric fields thatcould accelerate particles to energies of 2 × 1019

B100 eV. High-energy protons could emit TeV pho-tons as curvature radiation (22), and high-energyelectrons as Inverse Compton emission (23). Suchmodels could be vindicated by the detection of en-ergy spectra, which are inconsistent with originat-ing from shock accelerated particles. An examplefor the latter would be very hard energy spectrawhich require high minimum Lorentz factors of ac-celerated particles.

The improved data from next-generation gamma-ray experiments can be compared with improvednumerical results. The latter have recently madevery substantial progress. General RelativisticMagnetohydrodyamic codes are now able to testmagnetic models of jet formation and accelera-tion (see the review by (71)). The Relativistic-Particle-in-Cell technique opens up the possibil-ity of greatly improving our understanding a widerange of issues including jet bulk acceleration,electromagnetic energy transport in jets, and par-ticle acceleration in shocks and in magnetic recon-nection while incorporating the different radiationprocesses (27; 28; 29; 30).

Blazar observations would benefit from an in-creased sensitivity in the 100 GeV to 10 TeV en-ergy range to discover weaker sources and to sam-ple the energy spectra of strong sources on shorttime scales. A low energy threshold in the 10-40GeV range would be beneficial to avoid the effect

4

of intergalactic absorption that will be describedfurther below. Increased sensitivity at high en-ergies would be useful for measuring the energyspectra of a few nearby sources like M 87, Mrk 421,and Mrk 501 at energies≫10 TeV and to constrainthe effect of intergalactic absorption in the wave-length range above 10 microns. The interpretationof blazar data would benefit from dense temporalsampling of the light curves. Such sampling couldbe achieved with gamma-ray experiments locatedat different longitudes around the globe.

3. Cosmic rays from star-forming galaxies

More than 60% of the photons detected byEGRET during its lifetime were produced as aresult of interactions between cosmic rays (CRs)and galactic interstellar gas and dust. This dif-fuse radiation represents approximately 90% ofthe MeV-GeV gamma-ray luminosity of the MilkyWay (40). Recently H.E.S.S. reported the de-tection of diffuse radiation at TeV energies fromthe region of dense molecular clouds in the inner-most 200pc around the Galactic Center (41), con-firming the theoretical expectation that hadronicCRs could produce VHE radiation in their inter-action with atomic or molecular targets, throughthe secondary decay of π◦’s. Only one extragalac-tic source of diffuse GeV radiation was found byEGRET: the Large Magellanic Cloud, located atthe distance of ∼ 55 kpc (42). A simple re-scalingargument suggests that a putative galaxy withMilky-Way-like gamma-ray luminosity, located atthe distance of 1Mpc, would have a flux of approx-imately 2.5×10−8 cm−2 s−1 (> 100MeV), well be-low the detection limit of EGRET and ∼ 2×10−4

of the Crab Nebula flux (> 1 TeV), well belowthe sensitivity of VERITAS and H.E.S.S. Thus, anext-generation gamma-ray observatory with sen-sitivity at least an order of magnitude better thanVERITAS would allow the mapping of GeV-PeVcosmic rays in normal local group galaxies, such asM31, and study diffuse radiation from more dis-tant extragalactic objects if their gamma-ray lu-minosity is enhanced by a factor of ten or moreover that of the Milky Way.

Nearby starburst galaxies (SBG’s), such asNGC253, M82, IC342, M51 exhibit regions ofstrongly enhanced star formation and supernova(SN) explosions, associated with gas clouds whichare a factor of 102−105 more dense than the aver-age Milky Way gas density of ∼ 1 proton per cm3.This creates nearly ideal conditions for the emis-sion of intense, diffuse VHE radiation, assuming

that efficient hadronic CR production takes placein the sites of the SNR’s (i.e. that the galacticCR origin paradigm is valid) and in colliding OBstellar winds (43). In addition, leptonic gamma-ray production through inverse-Compton scatter-ing of high density photons produced by OB asso-ciations may become effective in star forming re-gions (44). Multiple attempts to detect SBGs havebeen undertaken by the first generation ground-based gamma-ray observatories. At TeV energies,M82, IC342, M81, and NGC3079 were observedby the Whipple 10m telescope (45), while M82and NGC253 were observed by HEGRA. How-ever, none of these objects were detected. A con-troversial detection of NGC 253 by the CANGA-ROO collaboration in 2002 (47) was ruled out byH.E.S.S. observations (48). The theoretical pre-dictions of TeV radiation from starburst galaxieshave not yet been confirmed by observations andthese objects will be intensively studied by the cur-rent generation instruments during the next sev-eral years. The optimistic theoretical considera-tions suggest that a few SBG’s located at distancesless than ∼ 10 Mpc may be discovered. Shouldthis prediction be confirmed, a next-generationgamma-ray observatory with sensitivity at least anorder of magnitude better than VERITAS will po-tentially discover thousands of such objects withinthe ∼ 100 Mpc visibility range. This will enablethe use of SBG’s as laboratories for the detailedstudy of the SNR CR acceleration paradigm andVHE phenomena associated with star formation,including quenching effects due to evacuation ofthe gas from star forming regions by SNR shocksand UV pressure from OB stars.

If accelerated CR’s are confined in the regions ofhigh gas or photon density long enough that theescape time due to diffusion through the magneticfield exceeds the interaction time, then the diffusegamma-ray flux cannot be further enhanced by anincreased density of target material, and insteadan increased SN rate is needed. Ultra LuminousInfraRed Galaxies (ULIRGs), which have SN rateson the scale of a few per year (compared to theMilky Way rate of ∼ 1 per century) and which alsohave very large amounts of molecular material,are candidates for VHE emission (49). Althoughlocated at distances between ten and a hundredtimes farther than the most promising SBG’s, theULIRG’s Arp220, IRAS17208, and NGC6240 maybe within the range of being detected by Fermi,VERITAS and H.E.S.S. (50). Next-generationgamma-ray instruments might be able to detectthe most luminous objects of this type even if they

5

are located at ∼ 1 Gpc distances. Initial studies ofthe population of ULIRGs indicate that these ob-jects underwent significant evolution through thehistory of the Universe and that at the moder-ate redshift (z < 1) the abundance of ULIRGsincreases. Any estimate of the number of ULIRGsthat may be detected is subject to large uncertain-ties due to both the unknown typical gamma-rayluminosity of these objects and their luminosityevolution. However, if theoretical predictions forArp220 are representative for objects of this type,then simple extrapolation suggests > 102 may bedetectable.

The scientific drivers to study ULIRG’s are simi-lar to those of SBGs and may include research ofgalaxy gamma-ray emissivity as a function of tar-get gas density, supernova rate, confining magneticfield, etc. In addition, research of ULIRGs mayoffer a unique possibility to observe VHE charac-teristics of star formation in the context of the re-cent history of the Universe (z < 1) since ULIRGsmight be detectable to much further distances.Other, more speculative, avenues of research mayalso be available. A growing amount of evidencesuggests that AGN feedback mechanism connectsepisodes of intense starbursts in the galaxies withthe accretion activity of central black holes. Onecan wonder then if a new insight into this phenom-ena can be offered by observation of VHE coun-terparts of these processes detected from dozensof ULIRGs in the range from 0.1-1 Gpc.

4. The largest particle accelerators in the

Universe: radio galaxies, galaxy clus-

ters, and large scale structure forma-

tion shocks

The possibility of observing diffuse GeV and TeVradiation from even more distant, rich galaxy clus-ters (GCs) has widely been discussed in the litera-ture. As the Universe evolves, and structure formson increasingly larger scales, the gravitational en-ergy of matter is converted into random kineticenergy of cosmic gas. In galaxy clusters, colli-sionless structure formation shocks, triggered byaccretion of matter or mergers, are thought to bethe main agents responsible for heating the inter-cluster medium (ICM) to temperatures of ∼ 10keV. Through these processes a fraction of grav-itational energy is converted into the kinetic en-ergy of non-thermal particles: protons and elec-trons. Galactic winds (51) and re-acceleration ofmildly relativistic particles injected into the ICMby powerful cluster members (52) may accelerate

additional particles to non-thermal energies. Cos-mic ray protons can escape clusters diffusively onlyon time scales much longer than the Hubble time.Therefore, they accumulate over the entire for-mation history (51) and interact with the inter-cluster thermal plasma to produce VHE gammaradiation. Theoretical predictions for the detec-tion of such systems in gamma rays by VERITASand H.E.S.S. include clusters in the range fromz = 0.01 to z = 0.25 (see Fig. 2) (43; 53; 54). Ob-jects of this category were observed with Whip-ple (55) and H.E.S.S. (56) but were not detected.Multiple attempts to find gamma-ray signals fromGCs in EGRET data also failed. Nevertheless,a large theoretical interest (58; 59; 60) motivatesfurther observations of the particularly promisingcandidates, such as the Coma and Perseus clus-ters by VERITAS and H.E.S.S.. If nearby repre-sentatives of the GC class are detected, a next-generation gamma-ray observatory with sensitiv-ity increased by a factor of 10 would be able toobtain spatially resolved energy spectra from theclose, high-mass systems, and should be able toobtain flux estimates and energy spectra of sev-eral dozen additional clusters. The detection ofgamma-ray emission from galaxy clusters wouldmake it possible to study acceleration mechanismson large scales (> 10 kpc). It would permit mea-surement of the energy density of non-thermal par-ticles and investigation of whether they affect theprocess of star formation in GCs, since their equa-tion of state and cooling behavior differs from thatof the thermal medium. If cosmic ray protons in-deed contribute noticeably to the pressure of theICM, measurements of their energy density wouldallow for improved estimates of the cluster massbased on X-ray data, and thus improve estimatesof the universal baryon fraction. Based on popu-lation studies of the gamma-ray fluxes from GCs,one could explore the correlation of gamma-ray lu-minosity and spectrum with cluster mass, temper-ature, and redshift. If such correlations are found,one could imagine using GCs as steady “standardcandles” to measure the diffuse infrared and visibleradiation of the Universe through pair-productionattenuation of gamma rays. From a theoreticalpoint of view the spectral properties of gamma-ray fluxes from GCs might be better understoodthan the intrinsic properties of blazars.

The anticipated discovery of extragalactic sourcesby VERITAS and H.E.S.S. will put theoreticalpredictions discussed here on firmer ground, atleast for the number of sources that the next gener-ation ground-based observatory may detect. Over

6

Fig. 2.—: Results from a cosmo-logical simulation showing how the >10 GeV gamma-ray emission from anearby rhich galaxy cluster could looklike when mapped with a gamma-raytelescope with 0.2◦ angular resolution.The image covers a 16◦ × 16◦ region(color scale: log(J/J) for an average>10 GeV flux of J = 8.2×10−9 cm−2

sec−1 sr−1) (from (53))

the next five years, Fermi will make major con-tributions to this area of studies. If the originof gamma radiation in these sources is hadronic,Fermi should be able to detect most of the SBGs,ULIRGs, and GCs, which could potentially be de-tected by VERITAS and H.E.S.S. Under some sce-narios, in which gamma rays are produced via lep-tonic mechanisms, a fraction of sources may es-cape Fermi detection (M82 might be such exam-ple), yet may still be detectable with VERITASand H.E.S.S. Future theoretical effort will be re-quired to guide observations of these objects. Ingeneral, benefiting from the full sky coverage ofFermi, a program to identify the Fermi sourcesusing the narrow field of view ACT observatoriesof the present day will be possible, and it is likelythat diffuse gamma-ray extragalactic sources willbe discovered. Fermi will measure the galacticand extragalactic gamma-ray backgrounds withunprecedented accuracy and will likely resolve themain contributing populations of sources in the en-ergy domain below a few GeV. The task of deter-mining the contribution from the diffuse gamma-ray sources to the extragalactic background in therange above a few GeV to ∼ 100 GeV will be bestaccomplished by the next generation ground-basedinstrument, capable of detecting a large number ofsources rather than a few. Most of these sourcesare anticipated to be weak, so they will requiredeep observations.

Large scale structure formation shocks could accel-erate protons and high-energy electrons out of the

intergalactic plasma. Especially in the relativelystrong shocks expected on the outskirts of clustersand on the perimeters of filaments, PeV electronsmay be accelerated in substantial numbers. CMBphotons Compton scattered by electrons of thoseenergies extend into the TeV gamma-ray spec-trum. The energy carried by the scattered pho-tons cools the electrons rapidly enough that theirrange is limited to regions close to the acceleratingshocks. However, simulations have predicted thatthe flux of TeV gamma rays from these shocks canbe close to detection limits by the current gener-ation of ground-based gamma-ray telescopes (61).If true, this will be one of the very few ways inwhich these shocks can be identified, since verylow thermal gas densities make their X-ray detec-tion virtually impossible. Since, despite the lowgas densities involved, these shocks are thought tobe a dominant means of heating cluster gas, theirstudy is vital to testing current models of cosmicstructure formation.

The origin of ultra-high-energy cosmic rays (UHE-CRs, E>

∼1016 eV) is one of the major unsolved

problems in contemporary astrophysics. Recently,the Auger collaboration reported tentative evi-dence for a correlation of the arrival directionsof UHECRs with the positions of nearby ActiveGalactic Nuclei. Gamma-ray observations may beideally suited to study the acceleration process, asthe UHECRs must produce gamma rays throughvarious processes. The UHECRs may be accel-erated far away from the black hole where the

7

Fig. 3.—: Fluxes from the electromagnetic cascade initi-ated in Cyg A by UHECRs assuming the total injectionpower of secondary UHE electrons and gamma rays in-jected at ≤ 1Mpc distances about 1045 erg/s. The solidand dashed lines show the synchrotron and Compton fluxes,respectively.

kpc jet is slowed down and dissipates energy. Ifthey are accelerated very close to the black holeat ∼pc distances, the high-energy particle beam isexpected to convert into a neutron beam throughphotohadronic interactions (63). On a length scalel ∼ 100 (En/1019 eV) kpc the neutron beam wouldconvert back into a proton beam through beta de-cays.

The interaction of UHECR with photons fromthe Cosmic Microwave Background (CMB) createssecondary gamma rays and electrons/positrons.Depending on the strength of the intergalac-tic magnetic field (BIGMF), a next-generationground-based gamma-ray experiment could detectGeV/TeV gamma rays from synchrotron emissionof first generation electrons/positrons (BIGMF ≥

10−9 G), or inverse Compton radiation froman electromagnetic cascade (BIGMF ≤ 10−9 G)(64). Figure 3 shows gamma-ray fluxes expectedfrom the electromagnetic cascade initiated in theCMBR and B = 3 µG environment of Cyg A byinjecting 1045 erg/s of secondary electrons and/orgamma rays from GZK protons. For the distanceto Cyg A of ≃ 240 Mpc the assumed radial sizeof the cluster R <∼ 1 Mpc corresponds to an ex-tended source, or halo, of angular size <∼ 14 ar-cmin. Although the absorption in EBL at TeV en-ergy is significant, the source should be detectablewith a next-generation experiment because thesource spectrum is very hard owing to synchrotronemission of UHE electrons. The detection of suchemission could give information about the ≫TeVluminosity of these sources, about the intensityand spectrum of the EBL, and about the strength

of the IGMF. A few aspects will be discussed fur-ther below.

A next-generation experiment might also be ableto detect gamma-ray haloes with diameters of afew Mpc around superclusters of galaxies. Suchhaloes could be powered by all the sources in thesupercluster that accelerate UHECRs. The sizeof the halo in these cases will be defined by thecombination of gyroradius of the UHE electronsand their cooling path (synchrotron and Comp-ton in Klein-Nishina regime). The spectral andspatial distibutions of such halos will contain cru-cial information about the EBL and intergalacticmagnetic fields.

5. Extragalactic radiation fields and extra-

galactic magnetic fields

Very high-energy gamma-ray beams traveling overextragalactic distances are a unique laboratory forstudying properties of photons, to constrain the-ories that describe spacetime at the Planck scaleand for testing radiation fields of cosmological ori-gin. The potential for probing the cosmic infraredbackground with TeV photons was first pointedout by Gould and Schreder (73) and was revivedby Stecker, de Jager & Salamon (11), inspiredby the detection of extragalactic TeV gamma-raysources in the nineties. High-energy gamma raystraveling cosmological distances are attenuated enroute to Earth by γ + γ → e+ + e− interac-tions with photons from the extragalactic back-ground light. While the Universe is transparentfor gamma-ray astronomy with energies below 10GeV, photons with higher energy are absorbed bydiffuse soft photons of wavelengths short enoughfor pair production. Photons from the EBL inthe 0.1 to 20 micron wavelength range renderthe Universe opaque in TeV gamma rays, simi-larly to the cosmic microwave background thatconstitutes a barrier for 100 TeV photons. Thetransition region from an observational windowturning opaque with increasing gamma-ray en-ergy provides the opportunity for deriving obser-vational constraints to the intervening radiationfield. Whereas the cosmic microwave backgroundis accessible via direct measurements, the cos-mic infrared background (CIB) has been elusiveand remains extremely difficult to discern by di-rect measurements. Energy spectra of extragalac-tic gamma-ray emitters between 10 GeV to 100TeV allow us to extract information about thediffuse radiative background using spectroscopicmeasurements. Non-thermal gamma-ray emission

8

spectra often extend over several orders of mag-nitude in energy and the high-energy absorptionfeatures expected from pair production can be ad-equately resolved with the typical energy resolu-tion of 10% to 20% achievable with atmosphericCherenkov telescopes.

The EBL, spanning the UV to far-infrared wave-length region, consists of the cumulative energyreleases in the Universe since the epoch of re-combination (see (74) for a review). The EBLspectrum comprises of two distinct components.The first, peaking at optical to near-infrared wave-lengths (0.5-2 µm), consists of primary redshiftedstellar radiation that escaped the galactic environ-ment either directly or after scattering by dust. Ina dust-free Universe, the SED of this componentcan be simply determined from knowledge of thespectrum of the emitting sources and the cosmichistory of their energy release. In a dusty Uni-verse, the total EBL intensity is preserved, butthe energy is redistributed over a broader spec-trum, generating a second component consistingof primary stellar radiation that was absorbed andreradiated by dust at infrared (IR) wavelengths.This thermal emission component peaks at wave-lengths around 100 to 140 µm. The EBL spectrumexhibits a minimum at mid-IR wavelengths (10 -30 µm), reflecting the decreasing intensity of thestellar contribution at the Rayleigh-Jeans part ofthe spectrum, and the paucity of very hot dustthat can radiate at these wavelengths.

All energy or particle releases associated with thebirth, evolution, and death of stars can ultimatelybe related to or constrained by the intensity orspectral energy distribution (SED) of the EBL.The energy output from AGN represent a majornon-nuclear contribution to the radiative energybudget of the EBL. Most of the radiative outputof the AGN emerges at X-ray, UV, and opticalwavelengths. However, a significant fraction of theAGN output can be absorbed by dust in the torussurrounding the accreting black hole, and reradi-ated at IR wavelengths. In addition to the radia-tive output from star forming galaxies and AGN,the EBL may also harbor the radiative imprint ofa variety of ”exotic” objects including PopulationIII stars, decaying particles, and primordial mas-sive objects. EBL measurements can be used toconstrain the contributions of such exotic compo-nents.

Direct detection and measurements of the EBLare hindered by the fact that it has no distinc-

tive spectral signature, by the presence of strongforeground emission from the interplanetary (zo-diacal) dust cloud, and from the stars and inter-stellar medium of the Galaxy. Results obtainedfrom TeV gamma-ray observations will comple-ment the results from a number of NASA mis-sions, i.e. Spitzer, Herschel, the Wide-Field In-frared Survey Explorer (WISE), and the JamesWebb Space Telescope (JWST). In order to de-rive the EBL density and spectrum via gamma-rayabsorption, ideally one would use an astrophysicalstandard candle of gamma rays to measure the ab-sorption component imprinted onto the observedspectrum. In contrast, extragalactic TeV gamma-ray sources detected to date are highly variableAGN. Their gamma-ray emission models are notunanimously agreed upon, making it impossible topredict the intrinsic source spectrum. Therefore,complementary methods are required for a con-vincing detection of EBL attenuation. Various ap-proaches have been explored to constrain/measurethe EBL intensity (11; 75; 76; 77; 78; 79), rang-ing from searching for cutoffs, the assumption ofplausible theoretical source models, the possibil-ity of using contemporaneous X-ray to TeV mea-surements combined with emission models and theconcept of simultaneous constraints from direct IRmeasurements/limits combined with TeV data viaexclusion of unphysical gamma-ray spectra. All ofthese techniques are useful; however, none has sofar provided an unequivocal result independent ofassumed source spectra.

The next-generation gamma-ray experiments willallow us to use the flux and spectral variabilityof blazars (80; 81; 82) to separate variable sourcephenomena from external persistent spectral fea-tures associated with absorption of the gamma-ray beam by the EBL. Redshift dependent studiesare required to distinguish possible absorption byradiation fields nearby the source from extragalac-tic absorption. The most prominent feature ofblazars is their occasional brightness (sometimes> 10 Crab) yielding a wealth of photon statistics.Those flares are to date the most promising testsof the EBL density based on absorption. To con-strain the EBL between the UV/optical all theway to the far IR a statistical sample of gamma-ray sources, and a broader energy coverage withproperly matched sensitivity are required.

Since the cross-section for the absorption of agiven gamma-ray energy is maximized at a specifictarget photon wavelength (e.g., a 1 TeV gamma-ray encounters a 0.7 eV soft photon with maxi-

9

mum cross-section), there is a natural division ofEBL studies with gamma rays into three regions:the UV to optical light, the near- to mid-IR andthe mid- to far-IR portion of the EBL are the mosteffective absorbers for ≈ 10 - 100 GeV, the ≈ 0.1TeV to 10 TeV and the ≈ 10 - 100 TeV regime,correspondingly.

In the search for evidence of EBL absorption inblazar spectra it is important to give considera-tion to the shape of the EBL spectrum showinga near IR peak, a mid IR valley and a far IRpeak; absorption could imprint different featuresonto the observed blazar spectra. For example, acutoff from the rapid increase of the opacity withgamma-ray energy and redshift is expected to bemost pronounced in an energy spectral regime thatcorresponds to a rising EBL density; e.g., as isfound between 0.1 - 2 micron. This correspondsto gamma ray energies of 10 GeV - 100 GeV. Asurvey with an instrument with sensitivity in the10 GeV to several 100s of GeV could measure acutoff over a wide range of redshifts and constrainthe UV/optical IR part of the EBL. Fermi, to-gether with existing ground-based telescopes, ispromising in yielding first indications or maybefirst conclusive results for a detection of the EBLabsorption feature. However, an instrument witha large collection area over the given energy rangeby using the ground-based gamma-ray detectiontechnique would allow stringent tests via spectralvariability measurements.

Similarly, a substantial rise in the opacity withgamma-ray energy is expected in the energyregime above 20 TeV, stemming from the far IRpeak. A corresponding cutoff should occur in the20-50 TeV regime. Prospective candidate objectsare Mrk 421, Mrk 501 or 1ES1959+650, as theyprovide episodes of high gamma-ray fluxes, allow-ing a search for a cutoff with ground-based instru-ments that have substantially enlarged collectionareas in 10 - 100 TeV regime. Sensitivity for de-tection of a cutoff in this energy regime requiresIACTs with a collection area in excess of 1km

2

.

Finally, a promising and important regime forground-based telescopes to contribute to EBL con-straints lies in the near and the mid IR (0.5 - 5micron). The peak in the near IR and the slope ofdecline in the mid IR could lead to unique spectralimprints onto blazar spectra around 1-2 TeV, as-suming sufficient instrumental sensitivity. A steepdecline could lead to a decrease in opacity, whereasa minimal decline could result in steepening of theslope of the source spectrum. If this feature is

sufficiently pronounced and/or the sensitivity ofthe instrument is sufficient, it could be a power-ful method in unambiguously deriving the level ofabsorption and discerning the relative near to midIR density. The location of the near IR peak and,consequently, the corresponding change in absorp-tion, is expected to occur around 1.5 TeV, whichrequires excellent sensitivity between 100 GeV and10 TeV. The discovery of a signature for EBL ab-sorption at a characteristic energy would be ex-tremely valuable in establishing the level of ab-sorption in the near to mid IR regime. The ori-gin of any signature could be tested using spectralvariations in blazar spectra and discerning a stablecomponent.

A powerful tool for studying the redshift depen-dence of the EBL intensity are pair haloes (83).For suitable IGMF strengths, such haloes will formaround powerful emitters of >100 TeV gammarays or UHECRs, e.g. AGN and galaxy clusters.If the intergalactic magnetic field (IGMF) is nottoo strong, the high-energy radiation will initi-ate intergalactic electromagnetic pair productionand inverse Compton cascades. For an intergalac-tic magnetic field (IGMF) in the range between10−12 G and 10−9 G the electrons and positronscan isotropize and can result in a spherical haloglowing predominantly in the 100 GeV – 1 TeVenergy range. These haloes should have large ex-tent with radial sizes > 1 Mpc. The size of a 100GeV halo surrounding an extragalactic source ata distance of 1 Gpc could be less than 3◦ and bedetectable with a next-generation IACT experi-ment. The measurement of the angular diameterof such a halo gives a direct estimate of the lo-cal EBL intensity at the redshift of the pair halo.Detection of several haloes would thus allow us toobtain unique information about the total amountof IR light produced by the galaxy populations atdifferent redshifts.

For a rather weak IGMF between ∼ 10−16 G and∼ 10−24 G, pair creation/inverse Compton cas-cades may create a GeV/TeV ”echo” of a TeVGRB or AGN flare (84). The IGMF may be dom-inated by a primordial component from quantumfluctuations during the inflationary epoch of theUniverse, or from later contributions by Popula-tion III stars, AGN, or normal galaxies. The timedelay between the prompt and delayed emissiondepends on the deflection of the electrons by theIGMF, and afford the unique possibility to mea-sure the IGMF in the above mentioned interval offield strengths.

10

REFERENCES

Buckley, J., Byrum, K., Dingus, B., Falcone, A.,

Kaaret, P., Krawczynski, H., Pohl, M., Vas-

siliev, V., Williams, D.A., “The Status and fu-

ture of ground-based TeV gamma-ray astron-

omy. A White Paper prepared for the Division

of Astrophysics of the American Physical Soci-

ety”, [arXiv:0810.0444].

Sreekumar, P., Bertsch, D. L., Hartman, R. C.,

Nolan, P. L. and Thompson, D. J., Astroparti-

cle Phys., 11, 221 (1999).

Aharonian, F. Akhperjanian, A. G., Bazer-Bachi,

A. R., et al., Science, 314, 1424 (2006)

Tavecchio, F., The Tenth Marcel Grossmann

Meeting, 512 (2005)

Krawczynski, H., Astronomical Society of the Pa-

cific Conference Series, 350, 105 (2006)

Hartman, R. C., et al., ApJS, 123, 79 (1999)

Punch, M., et al., Nature, 358, 477 (1992)

Fossati, G., et al., ApJ, 677, L2 (2008)

Aharonian, F., et al., A&A, 441, 465 (2005)

Horan, D., & Weekes, T. C., New Astronomy Re-

view, 48, 527 (2004)

Stecker, F. W., De Jager, O. C., & Salamon,

M. H.,ApJ, 390, L49 (1992)

Gaidos, J. A., et al., Nature, 383, 319 (1996)

Buckley, J. H., et al., ApJ, 472, L9 (1996)

Takahashi, T., et al., ApJ, 470, L89 (1996)

Unwin, S.C., Wehrle, A.E., Jones, D.L., Meier,

D.L., and Piner, B.G., Publ. of the Astronomi-

cal Society of Australia, 19, 5 (2002)

Grindlay, J.E., New Astron.Rev., 49, 436 (2005)

Krawczynski, H., Coppi, P. S., Maccarone, T., &

Aharonian, F. A., A&A, 353, 97 (2000)

Georganopoulos, M., Perlman, E. S., Kazanas, D.,

& McEnery, J., ApJ, 653, L5 (2006)

Blandford, R. D., MNRAS, 176, 465 (1976)

Blandford, R. D., & Znajek, R. L., MNRAS, 179,

433 (1997)

Lovelace, R. V. E., Nature, 262, 649 (1976)

Levinson, A., Physical Review Letters, 85, 912

(2000)

Krawczynski, H., ApJ, 659, 1063 (2007)

Romanova, M. M., & Lovelace, R. V. E. 1997,

ApJ, 475, 97 (1997)

Kundt, W., Gopal, K., JApA, 25, 115 (2004)

De Young, D., ApJ, 648, 200 (2006)

Noguchi, K., & Liang, E., Astrophys. & Space Sci.,

307, 315 (2007)

Nishikawa, K.-I., Hededal, C.B., Hardee, P.E.,

Fishman, G.J.,Kouveliotou, C., & Mizuno, Y.,

Astrophys. & Space Sci., 307, 319, (2007)

Lovelace, R.V.E., Gandhi, P.R., & Romanova,

M.M., Astrophys. & Space Sci., 298, 115 (2005)

Chang, P., Spitkovsky, A., & Arons, J., ApJ, in

press (arXiv:0704.3832v1) (2008)

Aharonian, F. A. 2000, New A, 5, 377 (2000)

Mucke, A., & Protheroe, R. J. Astroparticle

Physics, 15, 121 (2001)

Mucke, A., Protheroe, R. J., Engel, R., Rachen,

J. P., & Stanev, T., Astroparticle Physics, 18,

593 (2003)

Mannheim, K., A&A, 269, 67 (1993)

Lovelace, R. V. E., MacAuslan, J., & Burns,

M., AIP Conf. Proc. 56: Particle Acceleration

Mech. in Astrophys., 56, 399 (1979)

Burns, M. L., & Lovelace, R. V. E., ApJ, 262, 87

(1982)

Blandford, R. D., & Levinson, A., ApJ, 441, 79

(1995)

Levinson, A., & Blandford, R., ApJ, 449, 86

(1995)

Halzen, F., Hooper, D., Astropart.Phys., 23, 537

(2005)

Strong, A., Moskalenko, I. V. & Reimer, O., Proc.

26th Int. Cosmic Ray Conf. (Salt Lake), 4, 52

(1999).

Aharonian, F. et al. Nature, 439, 695 (2006).

Sreekumar, P. et al., ApJ, 400 L67 (1992).

Volk, H. J., Aharonian, F. A. & Breitschwerdt, D.,

Space Science Reviews 75, 279 (1996).

Pohl, M., A&A, 287, 453 (1994).

Nagai, T., Search for TeV Gamma-Ray Emission

from Nearby Starburst Galaxies (Ph.D. The-

sis), UMI Publishing (2005).

Gotting, N., Ph.D. Thesis (2006)

11

Itoh, C., et al., A&A, 396, L1 (2002).

Aharonian, F. et al. A&A, 442, 177 (2005).

Torres, D. F., Reimer, O., Domingo-Santamarıa,

E. & Digel, S. W., ApJ, 607, L99 (2004).

Torres, D. F., ApJ, 617, 966 (2004).

Volk, H. J., Atoyan, A. M., Astroparticle Physics,

11, 73 (1999),

Ensslin, T. A. & Biermann, P. L., A&A, 330, 90

(1998).

Keshet, U., Waxman, E., Loeb, A., Springel, V. &

Hernquist, L., ApJ, 585, 128 (2003).

Gabici, S. & Blasi, P., Astroparticle Physics, 20,

579 (2004)

Perkins, J. S. et al. ApJ, 644, 148, 2006

W. Domainko, W. Benbow, J. A. Hinton, et al.,

Contrib. to the 30th ICRC, Merida, Mexico

[arXiv:0708.1384] (2007)

Voelk, H. J., arXiv:astro-ph/0303078 (2003).

Berrington, R. C. & Dermer, C. D.

arXiv:astro-ph/0407278 (2004)

Reimer, A., Reimer, O., Schlickeiser, R. & Iyudin,

A., A&A, 424 773, (2004)

Rordorf, C., Grasso, D. & Dolag, K., Astroparticle

Physics, 22, 167 (2004)

Miniati, F. MNRAS, 342, 1009 (2003)

V. Van Elewyck, Proc. Int. School of Cosmic

Ray Astrophysics, 15th Course, Erice, Italy,

astro-ph/0612731 (2006)

A. Atoyan and C. D. Dermer 2003, Astrophys. J.,

586, 79

Gabici, S., & Aharonian, F. A., Physical Review

Letters, 95, 251102 (2005)

Atoyan, A., and Dermer, C., ApJ(in preparation)

(2007)

Aharonian, F., et al., ApJL, 664, L71 (2007)

Krawczynski, H., et al., ApJ, 559, 187 (2001)

Begelman, M. C., Fabian, A. C., and Rees, M. J.,

MNRAS, 384, L19 (2008)

Fossati, G., et al., ApJ, 677, 906 (2008)

Marscher, A. P., et al., Nature, 452, 966 (2008)

Spruit, H. C., ArXiv e-prints, 0804.3096 (2008)

Teshima, M., et al., ArXiv e-prints, 0709.1475

(2007)

Gould, R. J., & Schreder, G., Phys. Rev., 155,

1408 (1967)

Hauser, M.G., & Dwek, E., ARA & A, 39, 249

(2001)

Biller, S.D. et.al., ApJ, 445, 227 (1995)

Vassiliev, V.V., Astroppart. Phys., 12, 217 (2000)

Dwek, E., Krennrich, F. & Arendt, R., ApJ 634,

155 (2005)

Aharonian, F., et al., Nature, 440, 1018 (2006)

Mazin, D., & Raue, M., astro-ph/0701694 (2007)

Coppi, P. S., & Aharonian, F. A. 1999, Astropar-

ticle Physics, 11, 35 (1999)

Krennrich, F. et al, ApJ, 579, L9 (2002)

Aharonian, F., et al., A&A, 393, 89 (2002)

Aharonian, F. A., Coppi, P. S., & Voelk, H. J.,

ApJ, 423, L5 (1994)

Plaga, R. 1995, Nature, 374, 430

This 2-column preprint was prepared with the AAS LATEXmacros v5.2.

12